I recently came across some surprising results from New York City’s implementation of the nationally championed “Rapid Re-housing model” for addressing homelessness, so I wanted to look at how we might respond when a model doesn’t work as it was intended. In my first post on Wednesday, I provided background and an overview on the Rapid Re-housing model. Today I’ll assess the New York City results and consider potential next steps.
The New York City Results
Last month, Ralph Nunez, President and CEO of the Institute for Children, Poverty and Homelessness (ICPH) published a concise report entitled “Rapidly Rehousing Homeless Families: New York City—a Case Study.” In the report, Nunez analyzes the New York City government’s implementation of Rapid Re-housing strategy in its homelessness services over the last eight years. He concludes that Rapid Re-housing has actually pushed more New York families into homelessness and increased the costs of government services for the homeless overall.
Nunez writes: “Although designed to reduce family homelessness, rapid-rehousing policies have actually had the opposite effect in New York City. By offering rental subsidies to sheltered families, government actually stimulated homelessness. Numerous families that were living doubled-up or in substandard housing saw an opportunity to secure new housing and entered the shelter system to get places in line.” In essence, Nunez contends that the offer of increased rental subsidies through the new Rapid Re-housing program encouraged families to leave their current precarious housing situations and “get in line” at the shelter.
New York City’s data also shows a 179% increase in the recidivism rate (meaning the number of families who eventually came back to homeless shelters after being transitioned out) during the Rapid Re-housing program period. This increased recidivism meant increased cost and pressure on the New York City shelter system overall. None of these were intended effects of the Rapid Re-housing model. It appears that a model which has been lauded as an innovative solution to homelessness actually increased the amount of homeless people in New York City.
Where is it working and what’s different about New York?
If Rapid Re-housing failed in New York City, why is it promoted by national organizations and government agencies? Is Nunez simply the first person to analyze the results of the model? In fact, Rapid Re-housing has seen successes in many areas across the nation. The National Alliance to End Homelessness (NAEH)—a leader in the Rapid Re-housing movement through education and consulting—offers several examples of successful implementation of the model. In a training module about its outcomes, the NAEH reports, “Rapid re-housing has by far the highest percentage of exits to permanent housing of all the interventions [for homeless families including shelters and transitional housing] at 75%.” Areas from Alameda County, CA to the Tennessee Valley saw decreases in cost and recidivism after using Rapid Re-housing, directly contradicting the results in New York City. What’s going on here?
One hypothesis is that Rapid Re-housing works best in rural areas and small towns as opposed to large metropolises like New York City. Yet a thorough look at the NAEH’s success stories (click on houses in the above map) also contains evidence against this. The NAEH reports: “Philadelphia, Pennsylvania rapidly re-housed 648 homeless families. Only 11 households (1.7 percent) have had a subsequent homeless episode. […] Salt Lake City, Utah rapidly re-housed over 1,000 families in the last 31 months. […] Most of the families, 86 percent, did not have a subsequent homeless episode.” In response to these reports, Nunez still argues that no city faces the diversity and sheer volume of New York’s homeless population. Ultimately though, Nunez does not advocate for a complete elimination of the Rapid Re-housing model. Rather, he wants communities to carefully review the model’s use, stressing that Rapid Re-housing should not be viewed as a “one-size-fits-all” solution.
Another hypothesis about the comparative successes of Rapid Re-housing is that some implementations of the model have only accounted for short-term gains, not long-term solutions. The New York data is a particularly valuable window into Rapid Re-housing because—unlike many cities that have only recently begun using the model (including the town I just finished working for)—New York operated with a Rapid Re-housing strategy for a period of six years, from 2005-2011. This means that they’ve had ample time to watch the model take shape and to evaluate the after-effects once the program ended. The New York City results suggest that once Rapid Re-housing subsidies tapered off, both the homeless shelter recidivism rate and the average length of stay in shelters went up, and those rates are projected to increase further in the next two years.
Time will tell whether other areas see similar results if they end their own Rapid Re-housing programs. For now though, it’s important to state that Rapid Re-housing is, in theory, intended as a long-term solution to homelessness. Indeed, in support of the Rapid Re-housing model, Secretary of Housing and Urban Development (HUD), Shaun Donovan stated positively, “We can have the greatest impact on homelessness by helping people who have just fallen into homelessness quickly get back out – by rapidly finding long-term living situations for them.” Unfortunately, much of the data on Rapid Re-housing is quite recent and has not yet demonstrated the long-term effects of the program. When HUD reports that “87.7% of [Rapid Re-housing] program participants exited to permanent housing,” that doesn’t tell us how long those participants remained in permanent housing and with what types of subsidies.
Without substantial long-term data to compare to New York’s, I’m not ready to say that Rapid Re-housing is always harmful in the long run. However, Nunez’s report suggests that areas implementing the Rapid Re-housing model should continue to evaluate the housing situations of program participants after they exit, ideally for a few years. Rapid Re-housing providers should also maintain shared databases to track participants who may reenter the shelter system at a later date.
The central question that the New York City case study reveals is: How do we define success in addressing homelessness? Is it immediate and visible: getting homeless people out of the streets and shelters right now but potentially watching them trickle back a couple months later? Or is it longer-term: ensuring that families find stable housing and never reenter the shelter system? Is it monetary: spending the least amount of government money to permanently house the maximum amount of families? Or is it more comprehensive: getting a core group of families to that hard-to-define state of “stability” and economic security?
Addressing homelessness has always been a tricky business. The U.S. Department of Housing and Urban Development, which offered grant money from 2009-2012 to incentivize use of the Rapid Re-housing model, stated in its “Year 2 Summary” of the grant program: “Homelessness prevention is difficult to strategically target, and it is hard to measure its effect on reducing literal homelessness. On the other hand, rapid re-housing transitions people who are literally homeless into housing quickly. It directly decreases the overall number of homeless persons in shelters and on the streets.”
It would be tempting to say that HUD is playing a numbers game with its Rapid Re-housing model, focusing on removing homeless people from streets and shelters immediately, rather than truly ensuring that they never experience homelessness again. Yet Nunez could also be accused of playing the numbers game because his argument (see the quote at the beginning of this post) is that families previously living “doubled-up or in substandard housing” made themselves “homeless” by entering the shelter system when incentivized by Rapid Re-housing vouchers. That’s where the increase in numbers came from. One could argue that those New York City families were essentially homeless before they came to the shelter because they were not living in their own homes but rather sharing with friends or occupying nontraditional dwellings.
One clear way to evaluate success is by asking those who utilize the system what works for them. Were they better off sleeping on a friend’s couch? Was their time in the shelter productive? Did Rapid Re-housing subsidies help them to secure housing and eventually become financially independent? While I know that statistics and surveys are an essential assessment tool, an evaluation of the Rapid Re-housing model should also account for the practical effects on participants. This is particularly important given the normative analysis that necessarily comes with any sort of report from a specific organization, no matter how well intentioned.
Homeless people are some of the most marginalized members of our society. The lack of a home can put even the most basic goals of safety, health, education and economic autonomy on hold for families and individuals. It is with this understanding that the Rapid Re-housing model intervenes and gives people the chance to stabilize themselves in a home immediately (as I described in my first post about this topic). The New York City report should encourage Rapid Re-housing providers to carefully evaluate their results over the long term, recognizing that the model may work better in some locations than in others. On a broader scale, I see the report as a call to reexamine our definitions of success in the area of homelessness prevention and to listen attentively to the needs of those we serve.
If you’ve worked on Rapid Re-housing strategy or implementation, what do you see as its strengths and weaknesses? How do you respond to the New York City data?